Training AI Co-Scientists w/ Shashwat Goel
About Event
🔬 AI4Science on alphaXiv
🗓 Friday January 16th 2026 · 11AM PT
🎙 Featuring Shashwat Goel
💬 Casual Talk + Open Discussion
🎥 Zoom: Upon Registration
Description: Researchers from Meta Superintelligence Labs and academic institutions developed a scalable method to train language models as 'AI co-scientists' capable of generating high-quality research plans across diverse scientific domains. This approach leverages automated extraction of research goals and grading rubrics from scientific literature, combined with a self-grading reinforcement learning framework, resulting in finetuned models that human experts prefer 70% of the time and show 10-15% relative improvements in rubric satisfaction scores.
Check out the full paper here!
Whether you’re working on the frontier of LLMs or just curious about anything AI4Science, we’d love to have you there.